ViewVC Help
View File | Revision Log | Show Annotations | Download File | Root Listing
root/radiance/ray/doc/man/man1/rtrace.1
Revision: 1.29
Committed: Fri Jan 26 18:30:27 2018 UTC (7 years, 3 months ago) by rschregle
Branch: MAIN
Changes since 1.28: +4 -2 lines
Log Message:
Documented treatment of contrib photons as global under rpict/rtrace

File Contents

# User Rev Content
1 rschregle 1.29 .\" RCSid "$Id: rtrace.1,v 1.28 2016/05/18 12:45:55 rschregle Exp $"
2 greg 1.1 .TH RTRACE 1 10/17/97 RADIANCE
3     .SH NAME
4     rtrace - trace rays in RADIANCE scene
5     .SH SYNOPSIS
6     .B rtrace
7     [
8     .B options
9     ]
10     [
11     .B $EVAR
12     ]
13     [
14     .B @file
15     ]
16     .B octree
17     .br
18     .B "rtrace [ options ] \-defaults"
19     .SH DESCRIPTION
20     .I Rtrace
21     traces rays from the standard input through the RADIANCE scene given by
22     .I octree
23     and sends the results to the standard output.
24     (The octree may be given as the output of a command enclosed in quotes
25     and preceded by a `!'.)\0
26     Input for each ray is:
27    
28     xorg yorg zorg xdir ydir zdir
29    
30     If the direction vector is (0,0,0), a bogus record
31     is printed and the output is flushed if the
32     .I -x
33 greg 1.24 value is one or zero.
34 greg 1.1 (See the notes on this option below.)\0
35     This may be useful for programs that run
36     .I rtrace
37     as a separate process.
38     In the second form, the default values
39     for the options (modified by those options present)
40     are printed with a brief explanation.
41     .PP
42     Options may be given on the command line and/or read from the
43     environment and/or read from a file.
44     A command argument beginning with a dollar sign ('$') is immediately
45     replaced by the contents of the given environment variable.
46     A command argument beginning with an at sign ('@') is immediately
47     replaced by the contents of the given file.
48     Most options are followed by one or more arguments, which must be
49     separated from the option and each other by white space.
50     The exceptions to this rule are the boolean options.
51     Normally, the appearance of a boolean option causes a feature to
52     be "toggled", that is switched from off to on or on to off
53     depending on its previous state.
54     Boolean options may also be set
55     explicitly by following them immediately with a '+' or '-', meaning
56     on or off, respectively.
57     Synonyms for '+' are any of the characters "yYtT1", and synonyms
58     for '-' are any of the characters "nNfF0".
59     All other characters will generate an error.
60     .TP 10n
61     .BI -f io
62     Format input according to the character
63     .I i
64     and output according to the character
65     .I o.
66     .I Rtrace
67     understands the following input and output formats: 'a' for
68     ascii, 'f' for single-precision floating point,
69     and 'd' for double-precision floating point.
70     In addition to these three choices, the character 'c' may be used
71     to denote 4-byte floating point (Radiance) color format
72     for the output of values only
73     .I (\-ov
74     option, below).
75     If the output character is missing, the input format is used.
76     .IP
77     Note that there is no space between this option and its argument.
78     .TP
79     .BI -o spec
80     Produce output fields according to
81     .I spec.
82     Characters are interpreted as follows:
83     .IP
84     o origin (input)
85     .IP
86     d direction (normalized)
87     .IP
88     v value (radiance)
89     .IP
90 greg 1.14 V contribution (radiance)
91     .IP
92 greg 1.1 w weight
93     .IP
94 greg 1.10 W color coefficient
95 greg 1.7 .IP
96 greg 1.1 l effective length of ray
97     .IP
98     L first intersection distance
99     .IP
100 greg 1.2 c local (u,v) coordinates
101     .IP
102 greg 1.1 p point of intersection
103     .IP
104     n normal at intersection (perturbed)
105     .IP
106     N normal at intersection (unperturbed)
107     .IP
108     s surface name
109     .IP
110     m modifier name
111     .IP
112 greg 1.6 M material name
113     .IP
114 greg 1.9 ~ tilde (end of trace marker)
115 greg 1.8 .IP
116 greg 1.1 If the letter 't' appears in
117     .I spec,
118     then the fields following will be printed for every ray traced,
119     not just the final result.
120 greg 1.7 If the capital letter 'T' is given instead of 't', then all rays will
121     be reported, including shadow testing rays to light sources.
122 greg 1.1 Spawned rays are indented one tab for each level.
123 greg 1.9 The tilde marker ('~') is a handy way of differentiating the final ray
124     value from daughter values in a traced ray tree, and usually appears
125     right before the 't' or 'T' output flags.
126 greg 1.8 E.g.,
127 greg 1.9 .I \-ov~TmW
128     will emit a tilde followed by a tab at the end of each trace,
129     which can be easily distinguished even in binary output.
130 greg 1.1 .IP
131     Note that there is no space between this option and its argument.
132     .TP
133 greg 1.6 .BI -te \ mod
134 greg 1.1 Append
135 greg 1.6 .I mod
136 greg 1.1 to the trace exclude list,
137     so that it will not be reported by the trace option
138     .I (\-o*t*).
139     Any ray striking an object having
140 greg 1.6 .I mod
141 greg 1.1 as its modifier will not be reported to the standard output with
142     the rest of the rays being traced.
143 greg 1.7 This option has no effect unless either the 't' or 'T'
144     option has been given as part of the output specifier.
145 greg 1.6 Any number of excluded modifiers may be given, but each
146 greg 1.1 must appear in a separate option.
147     .TP
148 greg 1.6 .BI -ti \ mod
149 greg 1.1 Add
150 greg 1.6 .I mod
151 greg 1.1 to the trace include list,
152 greg 1.8 so that it will be reported by the trace option.
153 greg 1.1 The program can use either an include list or an exclude
154     list, but not both.
155     .TP
156     .BI -tE \ file
157     Same as
158     .I \-te,
159 greg 1.6 except read modifiers to be excluded from
160 greg 1.1 .I file.
161     The RAYPATH environment variable determines which directories are
162     searched for this file.
163 greg 1.6 The modifier names are separated by white space in the file.
164 greg 1.1 .TP
165     .BI -tI \ file
166     Same as
167     .I \-ti,
168 greg 1.6 except read modifiers to be included from
169 greg 1.1 .I file.
170     .TP
171     .BR \-i
172     Boolean switch to compute irradiance rather than radiance values.
173     This only affects the final result, substituting a Lambertian
174     surface and multiplying the radiance by pi.
175     Glass and other transparent surfaces are ignored during this stage.
176     Light sources still appear with their original radiance values,
177     though the
178     .I \-dv
179     option (below) may be used to override this.
180     This option is especially useful in
181     conjunction with ximage(1) for computing illuminance at scene points.
182     .TP
183 greg 1.13 .BR \-u
184     Boolean switch to control uncorrelated random sampling.
185 greg 1.12 When "off", a low-discrepancy sequence is used, which reduces
186     variance but can result in a brushed appearance in specular highlights.
187     When "on", pure Monte Carlo sampling is used in all calculations.
188     .TP
189 greg 1.1 .BR \-I
190     Boolean switch to compute irradiance rather than radiance,
191     with the input origin and direction interpreted instead
192     as measurement point and orientation.
193     .TP
194     .BR \-h
195     Boolean switch for information header on output.
196     .TP
197     .BI -x \ res
198     Set the x resolution to
199     .I res.
200     The output will be flushed after every
201     .I res
202 greg 1.21 input rays if
203     .I \-y
204     is set to zero.
205     A value of one means that every ray will be flushed, whatever
206     the setting of
207     .I \-y.
208 greg 1.1 A value of zero means that no output flushing will take place.
209     .TP
210     .BI -y \ res
211     Set the y resolution to
212     .I res.
213     The program will exit after
214     .I res
215     scanlines have been processed, where a scanline is the number of rays
216     given by the
217     .I \-x
218     option, or 1 if
219     .I \-x
220     is zero.
221     A value of zero means the program will not halt until the end
222     of file is reached.
223     .IP
224     If both
225     .I \-x
226     and
227     .I \-y
228     options are given, a resolution string is printed at the beginning
229     of the output.
230     This is mostly useful for recovering image dimensions with
231     .I pvalue(1),
232     and for creating valid Radiance picture files using the color output
233     format.
234     (See the
235     .I \-f
236     option, above.)
237     .TP
238 greg 1.18 .BI -n \ nproc
239     Execute in parallel on
240     .I nproc
241     local processes.
242 greg 1.19 This option is incompatible with the
243 greg 1.18 .I \-P
244     and
245     .I \-PP,
246 greg 1.19 options.
247 greg 1.18 Multiple processes also do not work properly with ray tree output
248     using any of the
249     .I \-o*t*
250     options.
251     There is no benefit from specifying more processes than there are
252     cores available on the system or the
253     .I \-x
254     setting, which forces a wait at each flush.
255     .TP
256 greg 1.1 .BI -dj \ frac
257     Set the direct jittering to
258     .I frac.
259     A value of zero samples each source at specific sample points
260     (see the
261     .I \-ds
262     option below), giving a smoother but somewhat less accurate
263     rendering.
264     A positive value causes rays to be distributed over each
265     source sample according to its size, resulting in more accurate
266     penumbras.
267     This option should never be greater than 1, and may even
268     cause problems (such as speckle) when the value is smaller.
269     A warning about aiming failure will issued if
270     .I frac
271     is too large.
272     .TP
273     .BI -ds \ frac
274     Set the direct sampling ratio to
275     .I frac.
276     A light source will be subdivided until
277     the width of each sample area divided by the distance
278     to the illuminated point is below this ratio.
279     This assures accuracy in regions close to large area sources
280     at a slight computational expense.
281     A value of zero turns source subdivision off, sending at most one
282     shadow ray to each light source.
283     .TP
284     .BI -dt \ frac
285     Set the direct threshold to
286     .I frac.
287     Shadow testing will stop when the potential contribution of at least
288     the next and at most all remaining light sources is less than
289     this fraction of the accumulated value.
290     (See the
291     .I \-dc
292     option below.)
293     The remaining light source contributions are approximated
294     statistically.
295     A value of zero means that all light sources will be tested for shadow.
296     .TP
297     .BI \-dc \ frac
298     Set the direct certainty to
299     .I frac.
300     A value of one guarantees that the absolute accuracy of the direct calculation
301     will be equal to or better than that given in the
302     .I \-dt
303     specification.
304     A value of zero only insures that all shadow lines resulting in a contrast
305     change greater than the
306     .I \-dt
307     specification will be calculated.
308     .TP
309     .BI -dr \ N
310     Set the number of relays for secondary sources to
311     .I N.
312     A value of 0 means that secondary sources will be ignored.
313     A value of 1 means that sources will be made into first generation
314     secondary sources; a value of 2 means that first generation
315     secondary sources will also be made into second generation secondary
316     sources, and so on.
317     .TP
318     .BI -dp \ D
319     Set the secondary source presampling density to D.
320     This is the number of samples per steradian
321     that will be used to determine ahead of time whether or not
322     it is worth following shadow rays through all the reflections and/or
323     transmissions associated with a secondary source path.
324     A value of 0 means that the full secondary source path will always
325     be tested for shadows if it is tested at all.
326     .TP
327     .BR \-dv
328     Boolean switch for light source visibility.
329     With this switch off, sources will be black when viewed directly
330     although they will still participate in the direct calculation.
331     This option is mostly for the program
332     .I mkillum(1)
333     to avoid inappropriate counting of light sources, but it
334     may also be desirable in conjunction with the
335     .I \-i
336     option.
337     .TP
338 greg 1.22 .BI -ss \ samp
339     Set the specular sampling to
340     .I samp.
341     For values less than 1, this is the degree to which the highlights
342     are sampled for rough specular materials.
343     A value greater than one causes multiple ray samples to be sent
344     to reduce noise at a commmesurate cost.
345 greg 1.1 A value of zero means that no jittering will take place, and all
346     reflections will appear sharp even when they should be diffuse.
347     .TP
348     .BI -st \ frac
349     Set the specular sampling threshold to
350     .I frac.
351     This is the minimum fraction of reflection or transmission, under which
352     no specular sampling is performed.
353     A value of zero means that highlights will always be sampled by
354     tracing reflected or transmitted rays.
355     A value of one means that specular sampling is never used.
356     Highlights from light sources will always be correct, but
357     reflections from other surfaces will be approximated using an
358     ambient value.
359     A sampling threshold between zero and one offers a compromise between image
360     accuracy and rendering time.
361     .TP
362     .BR -bv
363     Boolean switch for back face visibility.
364 greg 1.25 With this switch off, back faces of all objects will be invisible
365     to view rays.
366 greg 1.1 This is dangerous unless the model was constructed such that
367 greg 1.25 all surface normals face outward.
368 greg 1.1 Although turning off back face visibility does not save much
369     computation time under most circumstances, it may be useful as a
370     tool for scene debugging, or for seeing through one-sided walls from
371     the outside.
372     .TP
373     .BI -av " red grn blu"
374     Set the ambient value to a radiance of
375     .I "red grn blu".
376     This is the final value used in place of an
377     indirect light calculation.
378     If the number of ambient bounces is one or greater and the ambient
379     value weight is non-zero (see
380     .I -aw
381     and
382     .I -ab
383     below), this value may be modified by the computed indirect values
384     to improve overall accuracy.
385     .TP
386     .BI -aw \ N
387     Set the relative weight of the ambient value given with the
388     .I -av
389     option to
390     .I N.
391     As new indirect irradiances are computed, they will modify the
392     default ambient value in a moving average, with the specified weight
393     assigned to the initial value given on the command and all other
394     weights set to 1.
395     If a value of 0 is given with this option, then the initial ambient
396     value is never modified.
397     This is the safest value for scenes with large differences in
398     indirect contributions, such as when both indoor and outdoor
399     (daylight) areas are visible.
400     .TP
401     .BI -ab \ N
402     Set the number of ambient bounces to
403     .I N.
404 greg 1.26 This is the maximum number of diffuse bounces computed by the indirect
405     calculation. A value of zero implies no indirect calculation.
406     .IP
407 rschregle 1.27 This value defaults to 1 in photon mapping mode (see
408 greg 1.26 .I -ap
409 rschregle 1.27 below), implying that global photon irradiance is always computed via
410 greg 1.26 .I one
411 rschregle 1.27 ambient bounce; this behaviour applies to any positive number of ambient
412     bounces, regardless of the actual value specified. A negative value enables
413     a preview mode that directly visualises the irradiance from the global
414     photon map without any ambient bounces.
415 greg 1.1 .TP
416     .BI -ar \ res
417     Set the ambient resolution to
418     .I res.
419     This number will determine the maximum density of ambient values
420     used in interpolation.
421     Error will start to increase on surfaces spaced closer than
422     the scene size divided by the ambient resolution.
423     The maximum ambient value density is the scene size times the
424     ambient accuracy (see the
425     .I \-aa
426     option below) divided by the ambient resolution.
427     The scene size can be determined using
428     .I getinfo(1)
429     with the
430     .I \-d
431     option on the input octree.
432     .TP
433     .BI -aa \ acc
434     Set the ambient accuracy to
435     .I acc.
436     This value will approximately equal the error
437     from indirect illuminance interpolation.
438     A value of zero implies no interpolation.
439     .TP
440     .BI -ad \ N
441     Set the number of ambient divisions to
442     .I N.
443     The error in the Monte Carlo calculation of indirect
444     illuminance will be inversely proportional to the square
445     root of this number.
446     A value of zero implies no indirect calculation.
447     .TP
448     .BI -as \ N
449     Set the number of ambient super-samples to
450     .I N.
451     Super-samples are applied only to the ambient divisions which
452     show a significant change.
453     .TP
454     .BI -af \ fname
455     Set the ambient file to
456     .I fname.
457     This is where indirect illuminance will be stored and retrieved.
458     Normally, indirect illuminance values are kept in memory and
459     lost when the program finishes or dies.
460     By using a file, different invocations can share illuminance
461     values, saving time in the computation.
462     The ambient file is in a machine-independent binary format
463     which can be examined with
464     .I lookamb(1).
465     .IP
466     The ambient file may also be used as a means of communication and
467     data sharing between simultaneously executing processes.
468     The same file may be used by multiple processes, possibly running on
469     different machines and accessing the file via the network (ie.
470     .I nfs(4)).
471     The network lock manager
472     .I lockd(8)
473     is used to insure that this information is used consistently.
474     .IP
475     If any calculation parameters are changed or the scene
476     is modified, the old ambient file should be removed so that
477     the calculation can start over from scratch.
478     For convenience, the original ambient parameters are listed in the
479     header of the ambient file.
480     .I Getinfo(1)
481     may be used to print out this information.
482     .TP
483 greg 1.6 .BI -ae \ mod
484 greg 1.1 Append
485 greg 1.6 .I mod
486 greg 1.1 to the ambient exclude list,
487     so that it will not be considered during the indirect calculation.
488     This is a hack for speeding the indirect computation by
489     ignoring certain objects.
490     Any object having
491 greg 1.6 .I mod
492 greg 1.1 as its modifier will get the default ambient
493     level rather than a calculated value.
494 greg 1.6 Any number of excluded modifiers may be given, but each
495 greg 1.1 must appear in a separate option.
496     .TP
497 greg 1.6 .BI -ai \ mod
498 greg 1.1 Add
499 greg 1.6 .I mod
500 greg 1.1 to the ambient include list,
501     so that it will be considered during the indirect calculation.
502     The program can use either an include list or an exclude
503     list, but not both.
504     .TP
505     .BI -aE \ file
506     Same as
507     .I \-ae,
508 greg 1.6 except read modifiers to be excluded from
509 greg 1.1 .I file.
510     The RAYPATH environment variable determines which directories are
511     searched for this file.
512 greg 1.6 The modifier names are separated by white space in the file.
513 greg 1.1 .TP
514     .BI -aI \ file
515     Same as
516     .I \-ai,
517 greg 1.6 except read modifiers to be included from
518 greg 1.1 .I file.
519     .TP
520 greg 1.26 .BI -ap " file [bwidth1 [bwidth2]]"
521     Enable photon mapping mode. Loads a photon map generated with
522     .I mkpmap(1)
523     from
524     .I file,
525     and evaluates the indirect irradiance depending on the photon type
526     (automagically detected) using density estimates with a bandwidth of
527     .I bwidth1
528     photons, or the default bandwidth if none is specified (a warning will be
529     issued in this case).
530     .IP
531     Global photon irradiance is evaluated as part of the ambient calculation (see
532     .I \-ab
533     above), caustic photon irradiance is evaluated at primary rays, and
534     indirect inscattering in
535     .I mist
536 rschregle 1.29 is accounted for by volume photons. Contribution photons are treated as
537     global photons by
538     .I rtrace.
539 greg 1.26 .IP
540     Additionally specifying
541     .I bwidth2
542     enables bias compensation for the density estimates with a
543     minimum and maximum bandwidth of
544     .I bwidth1
545     and
546     .I bwidth2,
547     respectively.
548     .IP
549     Global photon irradiance may be optionally precomputed by
550     .I mkpmap(1),
551     in which case the bandwidth, if specified, is ignored, as the nearest photon
552     is invariably looked up.
553     .IP
554     Using direct photons replaces the direct calculation with density estimates
555     for debugging and validation of photon emission.
556     .TP
557     .BI -am " frac"
558 rschregle 1.27 Maximum search radius for photon map lookups. Without this option, an
559     initial maximum search radius is estimated for each photon map from the
560     average photon distance to the distribution's centre of gravity. It is then
561     adapted to the photon density in subsequent lookups. This option imposes a
562     global fixed maximum search radius for
563     .I all
564     photon maps, thus defeating the automatic adaptation. It is useful when
565     multiple warnings about short photon lookups are issued. Note that this
566     option does not conflict with the bandwidth specified with the
567     .I \-ap
568     option; the number of photons found will not exceed the latter, but may be
569     lower if the maximum search radius contains fewer photons, thus resulting in
570     short lookups. Setting this radius too large, on the other hand, may
571     degrade performance.
572 greg 1.26 .TP
573 rschregle 1.28 .BI -ac " pagesize"
574     Set the photon cache page size when using out-of-core photon mapping. The
575     photon cache reduces disk I/O incurred by on-demand loading (paging) of
576     photons, and thus increases performance. This
577     is expressed as a (float) multiple of the density estimate bandwidth
578     specified with
579     .I \-ap
580     under the assumption that photon lookups are local to a cache page. Cache
581     performance is sensitive to this parameter: larger pagesizes will reduce the
582     paging frequency at the expense of higher latency when paging does occur.
583     Sensible values are in the range 4 (default) to 16.
584     .TP
585     .BI -aC " cachesize"
586     Set the total number of photons cached when using out-of-core photon
587     mapping, taking into account the pagesize specified by
588     .I \-ac.
589     Note that this is approximate as the number of cache pages is rounded to
590     the nearest prime. This allows adapting the cache to the available physical
591     memory. In conjunction with the
592     .I \-n
593     option, this is the cache size
594     .I per parallel process.
595     Cache performance is less sensitive to this parameter,
596     and reasonable performance can obtained with as few as 10k photons. The
597     default is 1M. This option recognises multiplier suffixes (k = 1e3, M =
598     1e6), both in upper and lower case.
599     .TP
600 greg 1.1 .BI -me " rext gext bext"
601     Set the global medium extinction coefficient to the indicated color,
602     in units of 1/distance (distance in world coordinates).
603     Light will be scattered or absorbed over distance according to
604     this value.
605     The ratio of scattering to total scattering plus absorption is set
606     by the albedo parameter, described below.
607     .TP
608     .BI -ma " ralb galb balb"
609     Set the global medium albedo to the given value between 0\00\00
610     and 1\01\01.
611     A zero value means that all light not transmitted by the medium
612     is absorbed.
613     A unitary value means that all light not transmitted by the medium
614     is scattered in some new direction.
615     The isotropy of scattering is determined by the Heyney-Greenstein
616     parameter, described below.
617     .TP
618     .BI \-mg \ gecc
619     Set the medium Heyney-Greenstein eccentricity parameter to
620     .I gecc.
621     This parameter determines how strongly scattering favors the forward
622     direction.
623     A value of 0 indicates perfectly isotropic scattering.
624     As this parameter approaches 1, scattering tends to prefer the
625     forward direction.
626     .TP
627     .BI \-ms \ sampdist
628     Set the medium sampling distance to
629     .I sampdist,
630     in world coordinate units.
631     During source scattering, this will be the average distance between
632     adjacent samples.
633     A value of 0 means that only one sample will be taken per light
634     source within a given scattering volume.
635     .TP
636     .BI -lr \ N
637     Limit reflections to a maximum of
638 greg 1.20 .I N,
639     if N is a positive integer.
640 greg 1.11 If
641     .I N
642     is zero or negative, then Russian roulette is used for ray
643     termination, and the
644     .I -lw
645     setting (below) must be positive.
646     If N is a negative integer, then this sets the upper limit
647 greg 1.20 of reflections past which Russian roulette will be used.
648 greg 1.11 In scenes with dielectrics and total internal reflection,
649     a setting of 0 (no limit) may cause a stack overflow.
650 greg 1.1 .TP
651     .BI -lw \ frac
652     Limit the weight of each ray to a minimum of
653     .I frac.
654 greg 1.11 During ray-tracing, a record is kept of the estimated contribution
655     (weight) a ray would have in the image.
656     If this weight is less than the specified minimum and the
657     .I -lr
658     setting (above) is positive, the ray is not traced.
659     Otherwise, Russian roulette is used to
660     continue rays with a probability equal to the ray weight
661     divided by the given
662     .I frac.
663 greg 1.1 .TP
664     .BR -ld
665     Boolean switch to limit ray distance.
666     If this option is set, then rays will only be traced as far as the
667     magnitude of each direction vector.
668     Otherwise, vector magnitude is ignored and rays are traced to infinity.
669     .TP
670     .BI -e \ efile
671     Send error messages and progress reports to
672     .I efile
673     instead of the standard error.
674     .TP
675     .BR \-w
676     Boolean switch to suppress warning messages.
677     .TP
678     .BI \-P \ pfile
679     Execute in a persistent mode, using
680     .I pfile
681     as the control file.
682     Persistent execution means that after reaching end-of-file on
683     its input,
684     .I rtrace
685     will fork a child process that will wait for another
686     .I rtrace
687     command with the same
688     .I \-P
689     option to attach to it.
690     (Note that since the rest of the command line options will be those
691     of the original invocation, it is not necessary to give any arguments
692     besides
693     .I \-P
694     for subsequent calls.)
695     Killing the process is achieved with the
696     .I kill(1)
697     command.
698     (The process ID in the first line of
699     .I pfile
700     may be used to identify the waiting
701     .I rtrace
702     process.)
703     This option may be used with the
704     .I \-fr
705     option of
706     .I pinterp(1)
707     to avoid the cost of starting up
708     .I rtrace
709     many times.
710     .TP
711     .BI \-PP \ pfile
712     Execute in continuous-forking persistent mode, using
713     .I pfile
714     as the control file.
715     The difference between this option and the
716     .I \-P
717     option described above is the creation of multiple duplicate
718     processes to handle any number of attaches.
719     This provides a simple and reliable mechanism of memory sharing
720     on most multiprocessing platforms, since the
721     .I fork(2)
722     system call will share memory on a copy-on-write basis.
723 greg 1.26 .SH NOTES
724     Photons are generally surface bound (an exception are volume photons), thus
725     the ambient irradiance in photon mapping mode will be biased at positions
726     which do not lie on a surface.
727 greg 1.1 .SH EXAMPLES
728     To compute radiance values for the rays listed in samples.inp:
729     .IP "" .2i
730 greg 1.15 rtrace \-ov scene.oct < samples.inp > radiance.out
731 greg 1.1 .PP
732     To compute illuminance values at locations selected with the 't'
733     command of
734     .I ximage(1):
735     .IP "" .2i
736 greg 1.16 ximage scene.hdr | rtrace \-h \-x 1 \-i scene.oct | rcalc \-e '$1=47.4*$1+120*$2+11.6*$3'
737 greg 1.1 .PP
738     To record the object identifier corresponding to each pixel in an image:
739     .IP "" .2i
740 greg 1.16 vwrays \-fd scene.hdr | rtrace \-fda `vwrays \-d scene.hdr` \-os scene.oct
741 greg 1.1 .PP
742     To compute an image with an unusual view mapping:
743     .IP "" .2i
744 greg 1.17 cnt 480 640 | rcalc \-e 'xr:640;yr:480' \-f unusual_view.cal | rtrace
745 greg 1.16 \-x 640 \-y 480 \-fac scene.oct > unusual.hdr
746 greg 1.26 .PP
747     To compute ambient illuminance in photon mapping mode from a global photon
748     map global.pm via one ambient bounce, and from a caustic photon map
749     caustic.pm at sensor positions in samples.inp:
750     .IP "" .2i
751     rtrace -h -ov -ab 1 -ap global.pm 50 -ap caustic.pm 50 scene.oct <
752     samples.inp > illum.out
753 greg 1.1 .SH ENVIRONMENT
754     RAYPATH the directories to check for auxiliary files.
755     .SH FILES
756 greg 1.5 /tmp/rtXXXXXX common header information for picture sequence
757 greg 1.1 .SH DIAGNOSTICS
758     If the program terminates from an input related error, the exit status
759     will be 1.
760     A system related error results in an exit status of 2.
761     If the program receives a signal that is caught, it will exit with a status
762     of 3.
763     In each case, an error message will be printed to the standard error, or
764     to the file designated by the
765     .I \-e
766     option.
767     .SH AUTHOR
768     Greg Ward
769     .SH "SEE ALSO"
770 greg 1.26 getinfo(1), lookamb(1), mkpmap(1), oconv(1), pfilt(1), pinterp(1),
771 greg 1.23 pvalue(1), rpict(1), rcontrib(1), rvu(1), vwrays(1), ximage(1)